Skip to main content

Moving averages: SMA, EMA and their uses

Moving averages are the foundation of 90% of technical strategies. Understand the difference between SMA and EMA, and their proper uses.

Moving Averages (MA) are the simplest technical tool that exists — and probably the most used. They're the foundation of many other indicators (MACD, Bollinger Bands, ADX). Understanding what they do, their different flavors, and their practical uses is acquiring one of the most transferable TA building blocks.

The principle

A moving average is exactly what the name says: the average of closing prices over a certain number of candles. You "slide" this average through time, candle by candle, and you get a smooth curve following price in rounded fashion.

An MA on 20 periods (MA 20) on a 4h chart gives you, for each candle, the average of the last 20 4h candles — roughly the last 80 hours, or 3.5 days.

Immediate utility: smooth the noise. An average tells you "there's the recent trend, without the erratic oscillations".

SMA vs EMA

Several types of moving averages exist, but two dominate:

SMA — Simple Moving Average

Straightforward arithmetic mean. Each candle in the window weighs equally. An SMA 50 = (close₁ + close₂ + ... + close₅₀) / 50.

Advantages: simple, robust, insensitive to outliers.
Disadvantage: slow. A recent move weights the same as a move 50 candles ago. It reacts with delay to trend changes.

EMA — Exponential Moving Average

The exponential moving average weighs recent candles more heavily. The last candle has more importance than older ones, and this weight decays exponentially.

Advantages: more reactive, it hugs recent price better, detects trend changes sooner.
Disadvantage: more sensitive to noise, more false signals on short TFs.

Which to choose? The vast majority of modern technical traders use EMA because it reflects recent market behavior better — and in crypto, with high volatility, it makes a difference. But SMA remains useful for high TFs and structural analysis.

In DYOR, we favor EMA for columns and filters.

Periods that matter

There are "standard" periods everyone uses, and which are thus self-reinforcing (tons of traders place orders around them):

  • EMA 9 / EMA 20: very reactive, useful for tracking trend near real-time on short TFs.
  • EMA 50: the "fast average" of reference. Often serves as dynamic support in uptrend (and dynamic resistance in downtrend).
  • EMA 100: less used but sometimes useful as intermediate filter.
  • EMA 200: the queen of MAs. On 4h and 1D, it's the line everyone watches. Above EMA 200, the market is said to be in "structural bullish trend". Below, "structural bearish trend".

Main uses

1. Determine the trend

The most basic and useful rule: if price is above EMA 200, bullish trend. Below, bearish trend. This alone is a powerful filter: all your long entries only happen when price is above EMA 200, and vice versa for shorts.

This rule alone would prevent most bad counter-trend trades beginners take.

2. Dynamic supports and resistances

EMAs act as moving levels. In an uptrend, price tends to bounce on key EMAs (especially 20, 50, 200) when it pulls back. This is called an EMA retest, and it's a classic entry setup.

In a downtrend, the same EMAs act as ceilings: price bounces them then drops again.

3. The golden cross and death cross

These dramatic terms refer to two famous crossings:

  • Golden cross: EMA 50 crosses above EMA 200. Structural bullish signal.
  • Death cross: EMA 50 crosses below EMA 200. Structural bearish signal.

On the daily, these crossings often herald structural moves of weeks or months. Less useful on shorter TFs (too noisy).

Caution: a golden cross doesn't mean "buy immediately now". It means "market regime just changed, favor longs over the coming weeks". It's a strategic filter, not a tactical signal.

4. Measure price distance

When price moves far from its EMA 20 (or 50), it tends to return to it. This is the basis of mean reversion strategies: "price is 20% above EMA 50, high probability it drops back" — use with care, as crypto extremes can persist.

Classic mistake: chasing fast crossings

I constantly see beginners use EMA 9 / EMA 20 crossing as their main entry signal. On short TFs, it's a disaster: you'll get a crossing every 2-3 candles, half will be fakes, you'll lose to fees and stop hunts without understanding why.

Fast crossings can be useful as timing on an already-validated setup — never as an isolated signal.

The importance of slope

An EMA isn't just a line with a value — it also has a slope. And the slope is often more informative than the value itself:

  • EMA rising sharply: bullish trend underway, momentum present.
  • EMA flattening: exhaustion, trend losing strength.
  • EMA turning: possible regime change beginning.

Always watch the slope of EMA 50, not just its position.

Known pitfalls

Pitfall #1: overlay too many MAs. Three different MAs displayed simultaneously (EMA 20, 50, 200) is a good compromise. Beyond that, your chart becomes unreadable. Adding SMA 9, SMA 21, SMA 34, SMA 55 in "Fibonacci system" fashion, and you lose in clarity what you gain in complexity.

Pitfall #2: use them in a range. In a sideways market, MAs zigzag uselessly and generate contradictory signals constantly. Always combine with ADX to filter: MA only useful if ADX > 20.

Pitfall #3: treat them as absolute support. EMA 200 isn't a wall. It's broken regularly, especially on big news. It's indicative, not guaranteed.

In DYOR

DYOR displays EMAs 20, 50, 200 on charts, and several Trendscanner columns use these levels:

  • Distance to EMA 200: at what % gap is price from its EMA 200.
  • Price above/below EMA 50: simple trend filter.
  • Recent golden/death cross: filter to find coins that just changed regime.

Combine with ADX to eliminate false signals in ranges.

To go further

  • ADX — the indispensable filter to know if your MAs make sense;
  • MACD — built on EMAs, you'll understand MACD better if you master EMAs;
  • Support and resistance — EMAs as dynamic support/resistance.

Related reads